Populations in the rural developing world have just as many, if not more, information needs as people living in more affluent areas. But their constraints — intermittent power, intermittent connectivity, limited education, literacy and capital — make first-world approaches to accessing information systems inapplicable. Mobile phones are on the cusp of spurring an information revolution in such regions. Long battery life, wireless connectivity, solid-state memory, low price and immediate utility make this device better suited to rural conditions than a PC. However, current software on mobile phones makes them hard to use and to program.
In this dissertation, I present the design, implementation and evaluation of CAM — a mobile application framework designed to address the information needs of the rural developing world. Beginning with a two-month participatory design study, including users with varying levels of education, I propose some general guidelines for user interface and system design in this context. Motivated by these guidelines, I present CAM. CAM applications are accessed by capturing barcodes on paper forms using the mobile phone camera, or entering numeric strings with the keypad. Supporting one-step navigation, direct linkage to paper practices and offline multi-media interaction, CAM is uniquely adapted to rural device, user and infrastructure constraints. To demonstrate the usability of this framework, I implement and evaluate several distinct CAM-based applications (one of which has already been commercially deployed). I also provide preliminary motivation for fourteen other applications that could be implemented with the same, or similar, approach.